fix: improve reliability of Run integration test

Author: danabens

ada.log

@@ -0,0 +1 @@
+{"_aws" : {"Timestamp":1679517110000,"CloudWatchMetrics":[{"Namespace":"toolbox-python-migration-tools","Dimensions":[[],["Platform"],["Channel"],["Command"],["Version"],["Platform","Channel"],["Command","Channel"],["Version","Channel"],["ToolName","SubCommand"]],"Metrics":[{"Name":"Success","Unit":"Count"}]}]},"Success" : 1,"Version" : "1.0.1880.0","ToolName" : "","SubCommand" : "update","Platform" : "macOS-amd64","Channel" : "stable","Command" : "flake8","Channel" : "stable","Arch" : "amd64","Command" : "flake8","User" : "danabens","InvocationId" : "1-641b65b6-3130261744f6c82e8f971c80","Os" : "osx","Platform" : "macOS-amd64","Hostname" : "88665a415b08.ant.amazon.com","Version" : "1.0.1880.0"}

tests/integ/sagemaker/experiments/test_run.py

@@ -15,6 +15,7 @@
 import datetime
 import json
 import os
+import time
 
 import pytest
 
@@ -245,6 +246,12 @@ def test_run_from_local_and_train_job_and_all_exp_cfg_match(
                 sagemaker_session=sagemaker_session,
             )
 
+            # the above estimator has wait=True but the job TC could still be receiving updates
+            # after wait is complete resulting in run TC being updated, then when the above with
+            # statement is exited another update trial component call is made _sometimes_
+            # resulting in a ConflictException
+            time.sleep(3)
+
         _check_tc_status_when_exiting(
             trial_component_name=run._trial_component.trial_component_name,
             init_start_time=init_start_time,